I’m trying to fine tune a model, but no matter what data I use, it always result in timeout while the model is loading, so I turned to using the data format on LLM Finetuning.
Base Model: meta-llama/Meta-Llama-3-8B-Instruct
Parameter:
{
“auto_find_batch_size”: “false”,
“chat_template”: “zephyr”,
“disable_gradient_checkpointing”: “false”,
“evaluation_strategy”: “epoch”,
“merge_adapter”: “false”,
“mixed_precision”: “fp16”,
“optimizer”: “adamw_torch”,
“peft”: “true”,
“padding”: “right”,
“quantization”: “int4”,
“scheduler”: “linear”,
“use_flash_attention_2”: “false”,
“batch_size”: “2”,
“block_size”: “1024”,
“max_completion_length”: “128”,
“dpo_beta”: “0.1”,
“epochs”: “3”,
“gradient_accumulation”: “4”,
“lr”: “0.00003”,
“logging_steps”: “-1”,
“lora_alpha”: “32”,
“lora_dropout”: “0.05”,
“lora_r”: “16”,
“max_grad_norm”: “1”,
“model_max_length”: “2048”,
“max_prompt_length”: “128”,
“save_total_limit”: “1”,
“seed”: “42”,
“warmup_ratio”: “0.1”,
“weight_decay”: “0”,
“model_ref”: “meta-llama/Meta-Llama-3-8B-Instruct”,
“target_modules”: “all-linear”
}
Error Log:
ERROR | 2024-06-11 21:22:27 | autotrain.trainers.common:wrapper:120 - train has failed due to an exception: Traceback (most recent call last):
File “C:\Users\James\anaconda3\envs\autotrain\lib\site-packages\autotrain\trainers\common.py”, line 117, in wrapper
return func(*args, **kwargs)
File “C:\Users\James\anaconda3\envs\autotrain\lib\site-packages\autotrain\trainers\clm_main_.py”, line 38, in train
train_dpo(config)
File “C:\Users\James\anaconda3\envs\autotrain\lib\site-packages\autotrain\trainers\clm\train_clm_dpo.py”, line 19, in train
train_data, valid_data = utils.process_data_with_chat_template(config, tokenizer, train_data, valid_data)
File “C:\Users\James\anaconda3\envs\autotrain\lib\site-packages\autotrain\trainers\clm\utils.py”, line 448, in process_data_with_chat_template
train_data = train_data.map(
File “C:\Users\James\anaconda3\envs\autotrain\lib\site-packages\datasets\arrow_dataset.py”, line 602, in wrapper
out: Union[“Dataset”, “DatasetDict”] = func(self, *args, **kwargs)
File “C:\Users\James\anaconda3\envs\autotrain\lib\site-packages\datasets\arrow_dataset.py”, line 567, in wrapper
out: Union[“Dataset”, “DatasetDict”] = func(self, *args, **kwargs)
File “C:\Users\James\anaconda3\envs\autotrain\lib\site-packages\datasets\arrow_dataset.py”, line 3156, in map
for rank, done, content in Dataset._map_single(**dataset_kwargs):
File “C:\Users\James\anaconda3\envs\autotrain\lib\site-packages\datasets\arrow_dataset.py”, line 3517, in _map_single
example = apply_function_on_filtered_inputs(example, i, offset=offset)
File “C:\Users\James\anaconda3\envs\autotrain\lib\site-packages\datasets\arrow_dataset.py”, line 3416, in apply_function_on_filtered_inputs
processed_inputs = function(*fn_args, *additional_args, **fn_kwargs)
File “C:\Users\James\anaconda3\envs\autotrain\lib\site-packages\autotrain\trainers\clm\utils.py”, line 268, in apply_chat_template
example[“chosen”] = ast.literal_eval(example[“chosen”])
File “C:\Users\James\anaconda3\envs\autotrain\lib\ast.py”, line 64, in literal_eval
node_or_string = parse(node_or_string.lstrip(" \t"), mode=‘eval’)
File “C:\Users\James\anaconda3\envs\autotrain\lib\ast.py”, line 50, in parse
return compile(source, filename, mode, flags,
File “”, line 1
hi nice to meet you
^^^^
SyntaxError: invalid syntax
ERROR | 2024-06-11 21:22:27 | autotrain.trainers.common:wrapper:121 - invalid syntax (, line 1)